The form statistics everyone quotes, and where they came from.
9 numbers that appear on nearly every form-builder blog, followed hop by hop to the original. 1 has no source we could find, 1 has been replaced by its own author, and 7 are real sources that measured something smaller than the sentence now attached to them.
By Mohith Kumar, founder of chatform · Traced September 2026
| The claim | Verdict | What the source shows |
|---|---|---|
| Conversational forms have a 40% higher completion rate than traditional forms. | no source found | We could not find any study behind the 40% figure; the chains we followed end at vendor blog posts that cite each other or cite nothing. |
| Typeform's forms average a 47.3% completion rate, against an industry average of 21.5%. | real, but narrower | The 47.3% is Typeform's own figure for its own platform in 2023; the 21.5% "industry average" it is compared with is given no source or method. |
| Multi-step forms convert 86% better — multi-page forms convert at 13.85% against 4.53% for single-page. | real, but narrower | These are two different sources merged into one claim: an observational 2015 Formstack report, and a 2019 HubSpot survey of 173 marketers' self-reported rates. |
| HubSpot found that cutting form fields from 4 to 3 increased conversions by 50%. | real, but narrower | HubSpot's analysis was a correlation across 40,000+ landing pages, and its author described the effect of more fields as a slight decrease — nobody removed a field. |
| Inline validation reduces form abandonment by 22% and cuts completion time by 42%. | real, but narrower | The study was a 22-person lab usability test, and it measured success rate and errors — it never mentions abandonment. |
| 81% of people have abandoned a form after starting to fill it out. | real, but narrower | A 2018 self-reported survey found 81% of people had recently abandoned at least one online form; it did not measure abandonment of any particular form. |
| Expedia removed one form field and gained $12 million a year. | real, but narrower | It is an executive's anecdote from a 2010 interview, and the field was confusing rather than extra: people typed their bank's name into "Company", which then failed address checks. |
| Reducing a form from 11 fields to 4 increased conversions by 120%. | real, but narrower | It was a before-and-after comparison of one contact form over two periods, and the whole effect rests on 16 extra submissions. |
| 22% of shoppers abandon a checkout because it is too long or complicated. | outdated | Baymard's current survey puts it at 17% of US online shoppers, among those who were not just browsing; 22% is the previous round. |
Each one, traced.
The original source, what it says in its own words, and how it was measured.
Verdict: no source found
“Conversational forms have a 40% higher completion rate than traditional forms.”
We could not find any study behind the 40% figure; the chains we followed end at vendor blog posts that cite each other or cite nothing.
Versions of it say 25–40%, 30% or "up to 40%". One widely copied post credits it to Formstack's form report, but that report contains no such number. The closest real figure is Typeform's own platform average, below — which is first-party and compared against an unsourced baseline.
Verdict: real, but narrower
“Typeform's forms average a 47.3% completion rate, against an industry average of 21.5%.”
The 47.3% is Typeform's own figure for its own platform in 2023; the 21.5% "industry average" it is compared with is given no source or method.
“an average form completion rate of 47.3%, more than 25 percentage points higher than the industry average completion rate of 21.5%”
How it was measured: Over 2.6 million forms published on Typeform in 2023 and 568 million submissions.
A vendor measuring its own product against a baseline it does not cite. The release also does not define "completion rate", so the two numbers may not even measure the same thing.
Verdict: real, but narrower
“Multi-step forms convert 86% better — multi-page forms convert at 13.85% against 4.53% for single-page.”
These are two different sources merged into one claim: an observational 2015 Formstack report, and a 2019 HubSpot survey of 173 marketers' self-reported rates.
“Single Page: 4.53% / Multi-Page: 13.85%”
How it was measured: Formstack: data from more than 650,000 anonymised form users, comparing different forms rather than testing one form both ways. HubSpot's "86% higher": a survey of 173 marketers, fielded in 2019.
Neither is an experiment. Formstack compared forms that were different in many ways besides page count, eleven years ago; HubSpot's number is what marketers said their conversion rates were, not what was measured.
Verdict: real, but narrower
“HubSpot found that cutting form fields from 4 to 3 increased conversions by 50%.”
HubSpot's analysis was a correlation across 40,000+ landing pages, and its author described the effect of more fields as a slight decrease — nobody removed a field.
“as the number of form fields increases, conversion rates decrease slightly, but not as steeply as I expected”
How it was measured: Cross-sectional analysis of more than 40,000 HubSpot customer landing pages.
The "50%" arrived later, through a blog summary ("increased by almost half") and then a 2013 infographic. It reads a chart from an observational study as if it were an A/B test.
Source: HubSpot (Dan Zarrella), “Which types of form fields lower landing page conversions?”, 2010
Verdict: real, but narrower
“Inline validation reduces form abandonment by 22% and cuts completion time by 42%.”
The study was a 22-person lab usability test, and it measured success rate and errors — it never mentions abandonment.
“a 22% increase in success rates, a 22% decrease in errors made, a 31% increase in satisfaction rating, a 42% decrease in completion times”
How it was measured: Eye-tracking usability test run by Etre in London with 22 participants.
A careful, useful study — of how people fill in a form in a lab. "Abandonment" was added in retelling.
Source: A List Apart (Luke Wroblewski), “Inline Validation in Web Forms”, 2009
Verdict: real, but narrower
“81% of people have abandoned a form after starting to fill it out.”
A 2018 self-reported survey found 81% of people had recently abandoned at least one online form; it did not measure abandonment of any particular form.
“81% of people recently abandoned at least one online form”
How it was measured: An online survey of consumers; the release itself does not state the sample size.
Having abandoned one form, ever recently, is a very different thing from 81% of your visitors leaving. The same survey found security concerns (29%) cited more often than length (27%).
Verdict: real, but narrower
“Expedia removed one form field and gained $12 million a year.”
It is an executive's anecdote from a 2010 interview, and the field was confusing rather than extra: people typed their bank's name into "Company", which then failed address checks.
“deleted that field – overnight there was a step function [change], resulting in $12m of profit a year”
No data or method was published. The lesson it actually supports is about an ambiguous optional field breaking a payment step — not that shorter forms earn millions.
Verdict: real, but narrower
“Reducing a form from 11 fields to 4 increased conversions by 120%.”
It was a before-and-after comparison of one contact form over two periods, and the whole effect rests on 16 extra submissions.
“Contact form conversions increased 120% when the number of fields was reduced from 11 to 4”
How it was measured: October–November 2007 against April–May 2008; 184 and 219 form views; submissions went from 10 to 26.
Not an A/B test, and small enough that a handful of submissions either way changes the headline.
Verdict: outdated
“22% of shoppers abandon a checkout because it is too long or complicated.”
Baymard's current survey puts it at 17% of US online shoppers, among those who were not just browsing; 22% is the previous round.
“17% of US online shoppers have abandoned an order due to a 'too long / complicated checkout process'”
How it was measured: Survey of US online shoppers' reasons for abandoning an order, excluding those only browsing.
Baymard is one of the few genuinely rigorous sources in this list — it just keeps updating, and the internet does not. Its average documented cart abandonment rate is 70.22%.
Source: Baymard Institute, “Cart Abandonment Rate Statistics (last updated 22 September 2025)”, 2025
The numbers that do hold up.
If you need a statistic for a form, use one of these: primary sources with a stated method.
Of people who start a form, 55.5% finish on desktop and 47.5% on mobile; of people who view one, 37.2% and 31.3% submit it.
Zuko form analytics benchmark, read September 2026 — stages defined, broken implementations excluded.
An AI chatbot that reads an open-ended answer and probes when it is thin drew significantly higher engagement and significantly better answers — more informative, more relevant, more specific and clearer — than the same questions asked as an ordinary web survey.
Ziang Xiao et al., ACM Transactions on Computer-Human Interaction 27(3), 2020
People answering through a chat interface gave more differentiated answers and were less likely to satisfice — to pick whatever ends the question fastest — than people answering the same survey on the web.
People disclosed more, and managed the impression they were making less, when they believed they were talking to a computer rather than to a person operating it.
Gale M. Lucas et al., Computers in Human Behavior 37, 94–100, 2014
Letting the interviewer explain what a question means, rather than reading fixed wording and leaving interpretation to the respondent, sharply reduced error in the answers.
Michael F. Schober et al., Public Opinion Quarterly 61(4), 576–602, 1997
The same effect held in a live household telephone survey: clarifying ambiguous questions improved accuracy, at the cost of longer interviews.
Frederick G. Conrad et al., Public Opinion Quarterly 64(1), 1–28, 2000
17% of US online shoppers have abandoned an order because the checkout was too long or too complicated — a complaint about the form itself, not the price.
Reminders that changed their wording as the sequence went on raised the odds of a response by over 30% against reminders that repeated themselves.
Henry Sauermann et al., Research Policy 42(1), 273–286, 2013
People shown a task as already begun, rather than as not yet started, were about twice as likely to finish it — with exactly the same work left to do either way.
Joseph C. Nunes et al., Journal of Consumer Research 32(4), 504–512, 2006
What these mean for how a form should ask its questions is on why conversation works, and the practical version is how to reduce form abandonment.
Questions about form statistics.
What is the average form completion rate?
There is no single honest number. The best-documented benchmark we found is Zuko's, which defines its stages: across its data, 37.2% of desktop form views and 31.3% of mobile views end in a submission, and 55.5% (desktop) and 47.5% (mobile) of people who start a form finish it. Your own form's per-question drop-off is more useful than any average.
Do conversational forms really have a 40% higher completion rate?
We could not find a study behind that number; the chains end at vendor blog posts. What peer-reviewed research does show is narrower: chat-style surveys produce more differentiated answers (Kim, Lee and Gweon, CHI 2019), and probing thin answers makes them more informative (Xiao et al., 2020).
Does reducing form fields increase conversions?
Often, but the famous numbers overstate it. HubSpot's 40,000-page analysis found conversion falls only slightly as fields increase; the 120% case study rests on 16 extra submissions; and Expedia's $12M field was a confusing field, not an extra one. Remove fields whose answers change no decision, and test it on your own form.
What percentage of people abandon checkout because it is too long?
Baymard Institute's current survey, last updated September 2025, finds 17% of US online shoppers have abandoned an order because the checkout was too long or complicated. The widely quoted 22% is from its previous survey round.
Found a number we traced wrong, or an original we missed? The point of this page is to be correct — write to us and we will update it, with the date.